SGLT 2 Inhibitor Dan GLP 1 RA

RESEARCH
Sodium-glucose cotransporter protein-2 (SGLT-2) inhibitors and
BMJ: first published as 10.1136/bmj.m4573 on 13 January 2021. Downloaded from http://www.bmj.com/ on 10 October 2022 by guest. Protected by copyright.
glucagon-like peptide-1 (GLP-1) receptor agonists for type 2
diabetes: systematic review and network meta-analysis of
randomised controlled trials
Suetonia C Palmer,1 Britta Tendal,2 Reem A Mustafa,3,4 Per Olav Vandvik,5 Sheyu Li,6,7
Qiukui Hao,8 David Tunnicliffe,9 Marinella Ruospo,10 Patrizia Natale,9,10 Valeria Saglimbene,10
Antonio Nicolucci,11 David W Johnson,12 Marcello Tonelli,13 Maria Chiara Rossi,11
Sunil V Badve,14 Yeoungjee Cho,12 Annie-Claire Nadeau-Fredette,15 Michael Burke,16
Labib I Faruque,17 Anita Lloyd,17 Nasreen Ahmad,17 Yuanchen Liu,17 Sophanny Tiv,17
Tanya Millard,2 Lucia Gagliardi,18,19 Nithin Kolanu,20 Rahul D Barmanray,21 Rita McMorrow,22
Ana Karina Raygoza Cortez,23 Heath White,2 Xiangyang Chen,6 Xu Zhou,24 Jiali Liu,25
Andrea Flores Rodríguez,23 Alejandro Díaz González-Colmenero,23 Yang Wang,26 Ling Li,25
Surya Sutanto,27 Ricardo Cesar Solis,23 Fernando Díaz González-Colmenero,23
René Rodriguez-Gutierrez,23 Michael Walsh,28,29 Gordon Guyatt,4 Giovanni F M Strippoli9,10
For numbered affiliations see ABSTRACT adults with type 2 diabetes with follow up of 24 weeks
end of the article. OBJECTIVE or longer. Studies were screened independently
Correspondence to: To evaluate sodium-glucose cotransporter-2 (SGLT- by two reviewers for eligibility, extracted data, and
G F M Strippoli
2) inhibitors and glucagon-like peptide-1 (GLP-1) assessed risk of bias.
gfmstrippoli@gmail.com
(or @gstrip3 on Twitter: receptor agonists in patients with type 2 diabetes at MAIN OUTCOME MEASURES
ORCID 0000-0002-6936-0616) varying cardiovascular and renal risk. Frequentist random effects network meta-analysis was
Additional material is published DESIGN carried out and GRADE (grading of recommendations
online only. To view please visit
the journal online. Network meta-analysis. assessment, development, and evaluation) used
Cite this as: BMJ 2021;372:m4573 DATA SOURCES to assess evidence certainty. Results included
http://dx.doi.org/10.1136/bmj.m4573 Medline, Embase, and Cochrane CENTRAL up to 11 estimated absolute effects of treatment per 1000
Accepted: 16 October 2020 August 2020. patients treated for five years for patients at very low
risk (no cardiovascular risk factors), low risk (three
ELIGIBILITY CRITERIA FOR SELECTING STUDIES
or more cardiovascular risk factors), moderate risk
Randomised controlled trials comparing SGLT-2
(cardiovascular disease), high risk (chronic kidney
inhibitors or GLP-1 receptor agonists with placebo,
disease), and very high risk (cardiovascular disease
standard care, or other glucose lowering treatment in
and kidney disease). A guideline panel provided
oversight of the systematic review.
RESULTS
WHAT IS ALREADY KNOWN ON THIS TOPIC
764 trials including 421 346 patients proved
Although trial results conflict, sodium-glucose transporter 2 (SGLT-2) inhibitors eligible. All results refer to the addition of SGLT-2
and glucagon-like peptide-1 (GLP-1) receptor agonists probably reduce inhibitors and GLP-1 receptor agonists to existing
cardiovascular and kidney disease when added to other glucose lowering diabetes treatment. Both classes of drugs lowered
treatment in adults with type 2 diabetes all cause mortality, cardiovascular mortality, non-
Uncertainty remains about the relative and absolute benefits and harms of these fatal myocardial infarction, and kidney failure
drugs across all important outcomes in patients with type 2 diabetes at varying (high certainty evidence). Notable differences were
levels of cardiovascular and kidney disease risk found between the two agents: SGLT-2 inhibitors
reduced admission to hospital for heart failure more
WHAT THIS STUDY ADDS than GLP-1 receptor agonists, and GLP-1 receptor
Benefits shared by SGLT-2 inhibitors and GLP-1 receptor agonists include agonists reduced non-fatal stroke more than SGLT-2
reductions in all cause and cardiovascular mortality, myocardial infarction, inhibitors (which appeared to have no effect). SGLT-2
kidney failure, and serious hyperglycaemia (high certainty evidence) and lower inhibitors caused genital infection (high certainty),
body weight (low certainty) without incurring severe hypoglycaemia (high whereas GLP-1 receptor agonists might cause severe
certainty) gastrointestinal events (low certainty). Low certainty
evidence suggested that SGLT-2 inhibitors and GLP-1
The two drug classes appear to have similar relative effects on all cause and
receptor agonists might lower body weight. Little or no
cardiovascular mortality, myocardial infarction, kidney failure, health related
evidence was found for the effect of SGLT-2 inhibitors
quality of life, and serious hyperglycaemia
or GLP-1 receptor agonists on limb amputation,
Absolute benefits of these drugs vary substantially based on cardiovascular blindness, eye disease, neuropathic pain, or health
and renal disease risk profiles of patients with type 2 diabetes, as reflected in related quality of life. The absolute benefits of these
risk stratified BMJ Rapid Recommendations directly informed by this systematic drugs vary substantially across patients from low to
review very high risk of cardiovascular and renal outcomes
the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 1

RESEARCH
(eg, SGLT-2 inhibitors resulted in 3 to 40 fewer deaths remains unavailable. Moreover, several large scale
in 1000 patients over five years; see interactive trials published recently require updated evidence
decision support tool (https://magicevidence.org/ synthesis.5 7 8 16
match-it/200820dist/#!/) for all outcomes. We therefore conducted a systematic review and
CONCLUSIONS network meta-analysis evaluating the benefits and
In patients with type 2 diabetes, SGLT-2 inhibitors harms of SGLT-2 inhibitors and GLP-1 receptor
and GLP-1 receptor agonists reduced cardiovascular agonists in adults with type 2 diabetes. This review is
and renal outcomes, with some differences in conducted as part of the BMJ Rapid Recommendations
benefits and harms. Absolute benefits are determined project, a collaborative initiative from the MAGIC
by individual risk profiles of patients, with clear (Making GRADE the Irresistible Choice) Evidence
implications for clinical practice, as reflected in the Ecosystem Foundation (https://magicevidence.org/)
BMJ Rapid Recommendations directly informed by this and The BMJ. The aim of the initiative is to provide
systematic review. trustworthy practice guidelines within months of
SYSTEMATIC REVIEW REGISTRATION newly released evidence, underpinned by rigorous
PROSPERO CRD42019153180. evidence summaries. This systematic review is part of
a forthcoming BMJ Rapid Recommendations cluster
Introduction and a fuller version on MAGICapp (www.magicapp.
Diabetes affects half a billion people worldwide org). Although the network meta-analysis includes all
and accounted for 1.5 million deaths in 2016.1 classes of diabetes drugs, it focuses on these SGLT-2
Glucose lowering is a mainstay of treatment.2 In inhibitors and GLP-1 receptor agonists added to other
people with type 2 diabetes and higher risks of diabetes drugs, and in relation to one another.17
cardiovascular disease, several large scale randomised
controlled trials have investigated sodium-glucose Methods
cotransporter-2 (SGLT-2) inhibitors and glucagon- Protocol registration
like peptide-1 (GLP-1) receptor agonists and reported We registered the protocol for this systematic review
reductions in cardiovascular mortality and non-fatal with PROSPERO (CRD42019153180).
cardiovascular complications with both classes of
drug.3 4 The mortality reductions have not, however, Guideline panel involvement
proved consistent across trials, leaving clinicians and In accordance with the BMJ Rapid Recommendations
patients uncertain of the magnitude of benefits.5-8 process, a guideline panel that comprised content
Recommendations released in 2019 by the American experts, diabetologists, nephrologists, internal
Diabetes Association include using SGLT-2 inhibitors medicine physicians, primary care physicians,
and GLP-1 receptor agonists in the management of methodologists, and patients provided critical
diabetes for people with cardiovascular disease or oversight of the review. The panel reviewed the
kidney disease who have not reached their glycaemic protocol, identified the population, selected and
target goals.2 European Society of Cardiology ranked important patient outcomes, and recommended
guidelines in 2019 recommended SGLT-2 inhibitors baseline risks on which absolute treatment effects were
and GLP-1 receptor agonists in people with type 2 calculated.
diabetes and cardiovascular disease or a high risk of
cardiovascular disease.9 National Institute for Health Search strategy
and Care Excellence (NICE) guidance, updated in 2019, The literature search designed by an information
suggested a stepped approach to intensification of specialist was conducted in Medline, Embase, and the
treatment with diabetes drugs, considering metformin Cochrane Central Register of Controlled Trials from
as the first treatment with additions of dual and triple March 2016 to 11 August 2020 without language
treatment from several drug classes, including SGLT-2 restriction (appendix 1). The search from a previous
inhibitors.10 network meta-analysis provided records from database
Several published systematic reviews and meta- inception to March 2016.11
analyses have summarised glucose lowering
treatments for people with type 2 diabetes, including Study selection
a network meta-analysis of stepped intensification Two reviewers, working independently, screened
of treatment added to metformin and sulfonylurea.11 citations and evaluated full text records for eligible
Results from more recent meta-analyses report studies using Covidence systematic review software
reductions in mortality and selected cardiovascular (Veritas Health Innovation, Melbourne, VIC, Australia;
and renal events with both SGLT-2 inhibitors and https://www.covidence.org). A third reviewer (BT or
GLP-1 receptor agonists.12-15 A network meta-analysis SCP) resolved disagreements by consensus.
including all available glucose lowering treatments, Parallel group randomised controlled trials were
the full range of important outcomes, GRADE (grading eligible if they compared SGLT-2 inhibitors or GLP-
of recommendations assessment, development, 1 receptor agonists with one another or with other
and evaluation) certainty ratings, and estimates glucose lowering treatments, placebo, or standard
of absolute benefits and harms in patients with care in adults with type 2 diabetes. We included
varying risks of cardiovascular and kidney disease studies reporting outcomes at 24 weeks or longer.
2 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

RESEARCH
Treatment was given either as monotherapy or added estimates in every closed loop of evidence using node
to non-randomised background glucose lowering splitting approaches, and for the entire network using
management and other treatments. a design-by-treatment interaction model.20 We imputed
Using definitions of outcomes specified in individual missing standard deviations for continuous variables
randomised controlled trials, the review examined all when absent using standard deviations borrowed from
patient-important outcomes as defined by the guideline other similar randomised controlled trials.21 22
panel, regardless of whether evidence was available in The guideline panel recommended five baseline risk
the eligible trials: all cause mortality, cardiovascular categories to estimate absolute effects of treatment
mortality, non-fatal myocardial infarction, non- on cardiovascular and kidney outcomes in different
fatal stroke, kidney failure, admission to hospital for categories of patients, reflecting typical clinical
heart failure, severe hypoglycaemia, blindness, eye scenarios in practice. Patient categories were defined
disease requiring intervention, health related quality as very low risk (no or few (<3) cardiovascular risk
of life, body weight, amputation, neuropathic pain, factors), low risk (three or more cardiovascular risk
diabetic ketoacidosis, serious hyperglycaemia, genital factors), moderate risk (cardiovascular disease), high
infection, Fournier gangrene, severe gastrointestinal risk (chronic kidney disease (estimated glomerular
events, pancreatic cancer, and pancreatitis. Glycated filtration rate 45-75 mL/min per 1.73 m2 with
haemoglobin A1c was not identified as an outcome albuminuria >300 mg/g (30 mg/mmol) or estimated
important to patients by the guideline panel, although glomerular filtration rate 15-45 mL/min per 1.73 m2)),
it was included in the review because it is prioritised by and very high risk (cardiovascular disease and chronic
some decision makers. kidney disease)). The panel calculated absolute
treatment effects of the network estimates based on
Data extraction the event rates from the best sources of evidence,
For each eligible study, two reviewers independently including cohort studies, risk prediction equations, or
extracted the following: study characteristics (year the placebo arm in available trials.23-28 We estimated
of publication, country or countries, funding, baseline absolute risk per 1000 patients treated for
duration), population (setting, sample size, patient five years, extrapolating from shorter term data when
demographics, coexisting illnesses), description of necessary and assuming a constant annual risk of the
interventions (drug class, name, dose), and outcomes. selected outcome for each year up to five years. A single
Reviewers resolved disagreements by discussion or, if estimate was obtained for all outcomes and treatment
necessary, consultation with a third reviewer. comparisons, which was used in combination with the
baseline risk estimates to present results as absolute
Risk of bias assessment effects. We planned subgroup analysis if sufficient
Two reviewers independently assessed risk of bias, data were available according to trial duration, BMI,
with adjudication by a third reviewer (SCP) using the presence of high cardiovascular risk, and presence of
Cochrane tool for assessing risk of bias in randomised chronic kidney disease. All pairwise and network meta-
trials, which includes random sequence generation, analyses were performed with Stata 13 MP (StataCorp,
allocation concealment, blinding, missing outcome College Station, TX) using published routines.29
data, and selective reporting of outcomes.18 Each
domain was judged as low, unclear, or high risk of bias. Assessment of evidence certainty
Reviewers considered allocation concealment as low The GRADE approach provided the methods to
risk if there were no reported methods for allocation assess the certainty of the evidence for direct and
concealment and participants and investigators were network comparisons, using a non-contextualised
blind to treatment allocation. approach.30 31 Ratings of evidence certainty for
direct estimates included considerations of risk of
Data synthesis and analysis bias, inconsistency, indirectness, imprecision, and
For each direct comparison of two treatments, we publication bias.
conducted a frequentist pairwise meta-analysis The lower of the ratings from the two direct
using a restricted maximum likelihood estimation estimates forming the dominant first order loop
and reported, with corresponding 95% confidence provided the starting point for certainty ratings for
intervals, odds ratios for dichotomous outcomes, mean each indirect estimate, with further downward rating
differences for continuous outcomes (body weight for intransitivity,32 if present. The estimate that
and glycated haemoglobin A1c), and standardised provided the most information—direct or indirect—was
mean difference for health related quality of life.19 We the basis for the certainty of the network estimates.
assessed statistical heterogeneity using the I2 statistic If these estimates contributed similar amounts of
and funnel plots for evidence of small study effects in information, we chose the higher of the two certainty
analyses including 10 or more studies. judgments. If evidence of incoherence between direct
We conducted network meta-analysis using and indirect estimates existed, the certainty of the
frequentist methods with restricted maximum likelihood network estimate was rated down, and we used the
estimation to quantify network heterogeneity, assuming estimate with the highest certainty—direct or indirect—
a common heterogeneity estimate within a network. as the best estimate of the treatment effect. MAGICapp
We assessed agreement between direct and indirect (https://www.magicapp.org/) provided the platform

RESEARCH
23 185
haemoglobin A1c was 8.1% and BMI was 30.1.
Eligibility criteria included coronary artery disease
Records identiﬁed through database searching
or macrovascular disease in 35 trials,3 6 7 33-64 atrial
4867 fibrillation in 1 trial,65 heart failure in 9 trials,66-74
Duplicate records excluded chronic kidney disease and/or albuminuria in 37
trials,8 75-110 and high risk of cardiovascular or kidney
18 318 disease in 10 trials.4 5 16 111-117 Treatment with SGLT-
Records screened on title and abstract 2 inhibitors or GLP-1 receptor agonists was generally
added to background glucose lowering treatment.
15 993
Records excluded during screening Risk of bias
Appendix 3 presents the risk of bias in each trial. The
2325 key limitation was low levels of reported blinding for
Full text articles assessed for eligibility participants, investigators, and outcome assessors.
Three hundred and seven (40.2%) of 764 trials were
1561 at low risk of bias in random sequence generation
Full text articles excluded and 529 (69.2%) were at low risk of bias in allocation
48 Incorrect patient population concealment. Four hundred and sixty two trials
363 Incorrect study design (60.5%) reported blinding for participants and
44 Incorrect intervention or comparator
912 Secondary publication of included trial
investigators, and 105 trials (13.7%) reported blinding
17 Retracted or terminated for outcome assessment. Three hundred and twenty
162 Ongoing trial without reported data eight trials (42.9%) were adjudicated as being at low
15 Data not extractable for meta-analysis risk of attrition bias and 355 trials (46.5%) were at low
risk of bias from selective outcome reporting.
764
Randomised controlled trials included in review Outcomes
Appendix 4 presents the network plot for each
Fig 1 | PRISMA flow diagram of studies included in the review of glucose lowering
treatments for type 2 diabetes outcome. No evidence of global network inconsistency
or heterogeneity was found except for health related
quality of life (appendix 5), and no serious concerns
to develop tables of the GRADE summary of findings. of incoherence between direct and indirect evidence
We also provide the absolute estimates of effect and (appendix 6). Appendix 7 presents network estimates
associated uncertainties across risk groups, derived for each drug comparison for all outcomes. The expected
from best relative estimates of effect from the network absolute differences in treatment with SGLT-2 inhibitors
meta-analysis. These results are presented in evidence or GLP-1 receptor agonists compared with placebo and
summaries and decision aids developed in MAGICapp, each other are shown in table 1, table 2, table 3, table
including an interactive decision support tool for 4, and appendix 8. The interactive decision support tool
multiple treatment choices (https://magicevidence. at https://magicevidence.org/match-it/200820dist/#!/
org/match-it/200820dist/#!/). shows absolute estimates of effect for key outcomes in
patients with type 2 diabetes, from very low to very high
Patient and public involvement risk of cardiovascular and renal outcomes.
Four patient partners living with type 2 diabetes were
included in the guideline panel that oversaw the All cause and cardiovascular mortality
generation of this paper. Patient partners were recruited Two hundred and thirty eight trials including 290 662
from patient advocacy organisations and Cochrane Task patients reported all cause mortality. SGLT-2 inhibitors
Exchange. They were involved as full panel members in lowered all cause mortality compared with placebo
modifying the research question and identifying and (odds ratio 0.85 (95% confidence interval 0.79 to
prioritising patient-important outcomes. There was no 0.92); 3 fewer per 1000 in five years for very low
public participation in this project. risk patients (moderate certainty); 10 fewer for low
risk patients (high certainty); 18 fewer for moderate
Results risk patients (high certainty); 26 fewer for high risk
Description of included studies patients (high certainty); and 40 fewer for very high
The electronic search yielded 23 185 unique records. risk patients (high certainty); table 4). GLP-1 receptor
Screening and full text article analysis identified 764 agonists lowered all cause mortality compared with
trials including 421 346 patients (fig 1, appendix 2) placebo (0.88 (0.83 to 0.94); 2, 8, 13, 17, and 24 fewer
comparing 11 glucose lowering drugs, placebo, or per 1000 in five years for very low, low, moderate, high,
standard care. Figure 2 shows the network of treatment and very high risk patients respectively, moderate to
comparisons in available trials. The trial sample size high certainty; table 4).
ranged from 16 to 17 160. The median trial mean age SGLT-2 inhibitors and GLP-1 receptor agonists had
was 57.1 years, and the median proportion of men was similar effects on all cause mortality (0.95 (0.86 to
55.6%. At baseline, the median trial mean glycated 1.06), moderate to high certainty; table 3).

RESEARCH
DPP-4 inhibitor
(77 395)
SGLT-2 inhibitor
(52 827) Thiazolidinedione
(30 949)
GLP-1 receptor agonist
(66 196)
Sulfonylurea
(29 462)
Basal insulin
(8756)
Metformin
(11 958)
Basal bolus insulin

(1327)
Standard treatment
(11 304)
Bolus insulin
(1280)
Alpha glucosidase inhibitor Placebo

(6982) Glitinide
(4393) (117 939)
Fig 2 | Network plot of trials evaluating glucose lowering treatments for type 2 diabetes. Network shows the number
of participants assigned to each glucose lowering class with the size of each circle proportional to the number of
randomly assigned participants in the treatment comparisons (sample size for the specific treatment shown in
brackets). Line widths are proportional to the number of trials comparing the corresponding pair of treatments.
The most frequent drug comparison was DPP-4 inhibitors compared with placebo. DPP-4=dipeptidyl peptidase-4
inhibitors; GLP-1=glucagon-like peptide-1; SGLT-2=sodium-glucose cotransporter-2
One hundred and thirty five trials including low, low, moderate, high, and very high risk patients
226 701 patients reported cardiovascular mortality. respectively, moderate to high certainty; table 1), as
SGLT-2 inhibitors lowered cardiovascular mortality did GLP-1 receptor agonists (0.88 (0.80 to 0.96); 2,
compared with placebo (0.84 (0.76 to 0.92); 2, 7, 5, 9, 12, and 18 fewer per 1000 in five years for very
12, 16, and 24 fewer per 1000 in five years for very low, low, moderate, high, and very high risk patients
Table 1 | Summary of anticipated absolute differences comparing sodium-glucose cotransporter-2 inhibitor treatment with placebo treatment per 1000
patients with diabetes type 2 and with very low to very high cardiovascular risk, treated for five years
Non-fatal Hospital
All cause Cardiovascular myocardial admission for Diabetic Genital
Risk* mortality mortality infarction Non-fatal stroke Kidney failure heart failure ketoacidosis infection Body weight
3 fewer 2 fewer 4 fewer 0 more 2 fewer
1 fewer
(4 fewer to 2 (3 fewer to 1 (6 fewer to 1 (3 fewer to 4 (2 fewer to 1
Very low (1 fewer to 0)
fewer) fewer) fewer) more) fewer)
⊕⊕⊕
⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕
10 fewer 7 fewer 7 fewer 1 more 3 fewer 9 fewer
(15 fewer to 6 (11 fewer to 4 (12 fewer to 2 (6 fewer to 8 (4 fewer to 1 (11 fewer to 7
Low
fewer) fewer) fewer) more) fewer) fewer)
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕
1.92 kg lower
18 fewer 12 fewer 13 fewer 1 more 6 fewer 23 fewer 0 143 more
(2.23 lower
(25 fewer to 10 (18 fewer to 6 (21 fewer to 3 (11 fewer to 13 (9 fewer to 2 (28 fewer to 17 (1 fewer to 2 (119 more to
Moderate to 1.62 lower)
fewer) fewer) fewer) more) fewer) fewer) more) 170 more)
over 6 months
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕⊕
⊕⊕
High
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕
Very high
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕
*Risk categories represent the following patient populations: very low=no or less than three cardiovascular risk factors; low=three or more cardiovascular risk factors; moderate=cardiovascular
disease; high=chronic kidney disease (reduced glomerular filtration rate or macroalbuminuria); very high=cardiovascular disease and chronic kidney disease.
Certainty of the evidence for each estimate is shown: high certainty ⊕⊕⊕⊕; moderate certainty ⊕⊕⊕; low certainty ⊕⊕; very low certainty ⊕.

RESEARCH
Table 2 | Summary of anticipated absolute differences comparing glucagon-like peptide-1 receptor agonist treatment with placebo treatment per 1000
patients with diabetes type 2 and with very low to very high cardiovascular risk, treated for five years
Non-fatal Hospital Severe
All cause Cardiovascular myocardial admission for gastrointestinal
Risk* mortality mortality infarction Non-fatal stroke Kidney failure heart failure events Body weight
2 fewer 2 fewer 5 fewer
2 fewer 0 0
(3 fewer to 1 (3 fewer to 1 (7 fewer to 2
Very low (4 fewer to 0) (1 fewer to 0) (1 fewer to 0)
fewer) fewer) fewer)
⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕
⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕
8 fewer 5 fewer 4 fewer 9 fewer 2 fewer 2 fewer
Low cardiovascular (11 fewer to 4 (9 fewer to 2 (8 fewer to 1 (13 fewer to 4 (1 fewer to 3 (4 fewer to 1
risk factors fewer) fewer) fewer) fewer) fewer) more)
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕
145 kg lower
58 more (9 more (1.72 lower
Moderate to 142 more) to 1.18 lower)
fewer) fewer) fewer) fewer) fewer) more)
⊕⊕ over 6 months
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕
⊕⊕
High
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕
Very high
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕
respectively, moderate to high certainty; table 2). SGLT- as did GLP-1 receptor agonists (0.92 (0.85 to 0.99);
2 inhibitors and GLP-1 receptor agonists did not have 2, 4, 8, 9, 13 fewer per 1000 in five years for very
different effects on cardiovascular mortality (0.96 low, low, moderate, high, and very high risk patients
(0.84 to 1.09, moderate to high certainty; appendix 8). respectively, moderate to high certainty; appendix
8). SGLT-2 inhibitors and GLP-1 receptor agonists
Non-fatal myocardial infarction did not have different effects on non-fatal myocardial
Two hundred and eight trials including 265 921 infarction (0.95 (0.84 to 1.08); moderate to high
patients reported non-fatal myocardial infarction. certainty; appendix 8).
SGLT-2 inhibitors lowered odds of non-fatal
myocardial infarction compared with placebo (odds Non-fatal stroke
ratio 0.87 (95% confidence interval 0.79 to 0.97); 4, One hundred and seventy six trials including 261 434
7, 13, 14, and 21 fewer per 1000 in five years for very patients reported non-fatal stroke. SGLT-2 inhibitors
low, low, moderate, high, and very high risk patients had little or no effect on non-fatal stroke (odds ratio
respectively, moderate to high certainty; appendix 8), 1.01 (95% confidence interval 0.89 to 1.14); moderate
Table 3 | Summary of anticipated absolute differences comparing sodium-glucose cotransporter-2 inhibitor treatment with glucagon-like peptide-1
receptor agonist treatment per 1000 patients with diabetes type 2 and with very low to very high cardiovascular risk, treated for five years
Non-fatal Hospital
All cause Cardiovascular myocardial Non-fatal admission for Diabetic Serious Genital
Risk* mortality mortality infarction stroke Kidney failure heart failure Body weight ketoacidosis hyperglycaemia infection
1 fewer 0 fewer 1 fewer 5 more 1 fewer
0
(3 fewer to 1 (2 fewer to 1 (4 fewer to 2 (1 more to 10 (2 fewer to 1
Very low (1 fewer to 0)
more) more) more) more) fewer)
⊕⊕
⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕
(10 fewer to 4 (6 fewer to 4 (8 fewer to 4 (1 more to 19 (2 fewer to 2 (10 fewer to 4
Low
more) more) more) more) more) fewer)
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕⊕ 0.47 kg lower
5 more 158 more
6 fewer 3 fewer 5 fewer 16 more 1 fewer 18 fewer (0.09 lower 1 more
(2 fewer to 19 (64 more to
(17 fewer to 7 (11 fewer to 6 (15 fewer to 7 (2 more to 33 (5 fewer to 3 (25 fewer to to 0.85 lower) (0 to 3 more)
Moderate more) 299 more)
more) more) more) more) more) 11 fewer) over 6 months ⊕⊕⊕
⊕⊕⊕ ⊕⊕⊕⊕
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕
(24 fewer to (15 fewer to 8 (16 fewer to 8 (3 more to 36 (21 fewer to (32 fewer to
High
10 more) more) more) more) 13 more) 13 fewer)
⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕⊕ ⊕⊕⊕ ⊕⊕⊕⊕
Very high (37 fewer to (22 fewer to 12 (24 fewer to (4 more to 53 (34 fewer to (66 fewer to
16 more) more) 11 more) more) 20 more) 27 fewer)

RESEARCH
Table 4 | GRADE summary of findings to illustrate absolute effects based on cardiovascular and renal risk, for all cause mortality for sodium-glucose
cotransporter-2 (SGLT-2) inhibitors and glucagon-like peptide-1 (GLP-1) receptor agonists compared with placebo or each other
Anticipated absolute effects over five years Anticipated absolute Certainty in
Relative effect (odds Baseline Risk with effects (95% CI) over treatment
Comparison ratio (95% CI)) risk* Risk with control intervention five years effects (GRADE) Plain text summary
SGLT-2 inhibitor 0.85 (0.79 to 0.92) Very low Placebo: 20 per SGLT-2 inhibitor: 3 fewer per 1000 Moderate due to SGLT-2 inhibitor treatment probably
v placebo 1000 17 per 1000 (from 2 fewer to 4 indirectness reduces all cause mortality in
fewer) people with diabetes and few or no
cardiovascular risk factors
Low Placebo: 70 per SGLT-2 inhibitor: 10 fewer per 1000 High SGLT-2 inhibitor treatment reduces all
1000 60 per 1000 (from 6 fewer to 15 cause mortality in people with diabetes
fewer) and cardiovascular risk factors
Moderate Placebo: 120 per SGLT-2 inhibitor: 18 fewer per 1000 High SGLT-2 inhibitor treatment reduces all
fewer) and established cardiovascular disease
High Placebo: 170 per SGLT-2 inhibitor: 26 fewer per 1000 High SGLT-2 inhibitor treatment reduces all
fewer) and chronic kidney disease
Very high Placebo: 265 per SGLT-2 inhibitor: 40 fewer per 1000 High SGLT-2 inhibitor treatment reduces all
fewer) and established cardiovascular disease
and chronic kidney disease
GLP-1 receptor 0.88 (0.83 to 0.94) Very low Placebo: 20 per GLP-1 receptor 2 fewer per 1000 Moderate due to GLP-1 receptor agonist treatment
agonist v placebo 1000 agonist: 18 per (from 1 fewer to 3 indirectness probably reduces all cause mortality
1000 fewer) in people with diabetes and few or no
cardiovascular risk factors
Low Placebo: 70 per GLP-1 receptor 8 fewer per 1000 High GLP-1 receptor agonist treatment
1000 agonist: 62 per (from 4 fewer to 11 reduces all cause mortality in people
1000 fewer) with diabetes and cardiovascular risk
factors
Moderate Placebo: 120 per GLP-1 receptor 13 fewer per 1000 High GLP-1 receptor agonist treatment
1000 fewer) with diabetes and established
cardiovascular disease
High Placebo: 170 per GLP-1 receptor 17 fewer per 1000 High GLP-1 receptor agonist treatment
1000 fewer) with diabetes and chronic kidney
disease
Very high Placebo: 265 per GLP-1 receptor 24 fewer per 1000 High GLP-1 receptor agonist treatment
1000 fewer) with diabetes and established
cardiovascular disease and chronic
kidney disease
SGLT-2 inhibitor 0.95 (0.86 to 1.06) Very low GLP-1 receptor SGLT-2 inhibitor: 1 fewer per 1000 Moderate due to SGLT-2 inhibitor treatment and GLP-1
v GLP-1 receptor agonist: 18 per 17 per 1000 (from 1 more to 3 indirectness receptor agonist treatment probably
agonist 1000 fewer) have similar effects on all cause
mortality in people with diabetes and
few or no cardiovascular risk factors
Low GLP-1 receptor SGLT-2 inhibitor: 4 fewer per 1000 High SGLT-2 inhibitor treatment and GLP-1
agonist: 62 per 58 per 1000 (from 4 more to 10 recptor agonist treatment have similar
1000 fewer) effects on all cause mortality in people
with diabetes and cardiovascular risk
factors
Moderate GLP-1 receptor SGLT-2 inhibitor: 6 fewer per 1000 High SGLT-2 inhibitor treatment and GLP-1
agonist: 107 per 101 per 1000 (from 7 more to 17 receptor agonist treatment have
1000 fewer) similar effects on all cause mortality in
people with diabetes and established
cardiovascular disease
High GLP-1 receptor SGLT-2 inhibitor: 9 fewer per 1000 High SGLT-2 inhibitor treatment and GLP-1
agonist: 153 per 144 per 1000 (from 10 more to 24 receptor agonist treatment have similar
1000 fewer) effects on all cause mortality in people
with diabetes and chronic kidney
disease
Very high GLP-1 receptor SGLT-2 inhibitor: 13 fewer per 1000 High SGLT-2 inhibitor treatment and GLP-1
agonist: 241 per 228 per 1000 (from 16 more to 37 receptor agonist treatment have
1000 fewer) similar effects on all cause mortality in
people with diabetes and established
cardiovascular disease and chronic
kidney disease
GRADE=grading of recommendations assessment, development, and evaluation.
The point estimate of the absolute effect for GLP-1 receptor agonist treatment, obtained from GLP-1 receptor agonist treatment versus placebo, was used to calculate the absolute effect for SGLT-
2 inhibitors versus GLP-1 receptor agonists.

RESEARCH
to high certainty; appendix 8). GLP-1 receptor agonists (EQ-5D), the Diabetes Quality of Life, Treatment
reduced non-fatal stroke (0.84 (0.76 to 0.93); 5, 9, Satisfaction Status Scale, and the Impact of Weight
16, 17, and 25 fewer per 1000 in five years for very on Quality of Life. SGLT-2 inhibitors had uncertain
low, low, moderate, high, and very high risk patients effects on health related quality of life (standardised
respectively, moderate to high certainty; appendix 8). mean difference 0.14 (95% confidence interval −0.08
SGLT-2 inhibitors had higher odds of non-fatal stroke to 0.36); low certainty; appendix 8). GLP-1 receptor
than GLP-1 receptor agonists (1.20 (1.03 to 1.41); agonists might increase health related quality of life
moderate to high certainty; appendix 8). (0.13 (0.03 to 0.23); low certainty; appendix 8). No
evidence was found that SGLT-2 inhibitors and GLP-1
Kidney failure receptor agonists had different effects on health related
Thirty three trials including 98 284 patients reported quality of life (0.01 (−0.19 to 0.20); low certainty;
kidney failure, defined generally as estimated appendix 8).
glomerular filtration rate below 15 ml/min per 1.73
m2 or start of kidney replacement treatment. SGLT- Body weight
2 inhibitors reduced kidney failure (odds ratio 0.71 Four hundred and sixty nine trials including 226 361
(95% confidence interval 0.57 to 0.89); 1,3, 6, 25, patients reported body weight during a median follow-
and 38 fewer per 1000 in five years for very low, up of six months. SGLT-2 inhibitor treatment and GLP-
low, moderate, high, and very high risk patients 1 receptor agonist treatment might lower body weight
respectively, moderate to high certainty; appendix 8). (mean difference for SGLT-2 inhibitors −1.92 kg (95%
GLP-1 receptor agonists also reduced kidney failure confidence interval −2.23 to −1.62); low certainty;
(0.78 (0.67 to 0.92); 0, 2, 4, 19, and 29 fewer per 1000 mean difference for GLP-1 receptor agonists −1.45 kg
in five years for very low, low, moderate, high, and (−1.72 to −1.18); low certainty; appendix 8). SGLT-2
very high risk patients respectively, moderate to high inhibitors appeared to lower body weight to a greater
certainty; appendix 8). SGLT-2 inhibitors and GLP- extent than GLP-1 receptor agonists (−0.47 kg (−0.85
1 receptor agonists probably did not have different to −0.09); moderate certainty).
effects on kidney failure (0.91 (0.69 to 1.20); low to
moderate certainty). Glycated haemoglobin A1c
Six hundred and four trials including 242 745 patients
Admission to hospital for heart failure reported glycated haemoglobin A1c during a median
One hundred and forty nine trials including 242 361 follow-up of six months. SGLT-2 inhibitors (mean
patients reported admission to hospital for heart difference −0.60% (95% confidence interval −0.67
failure. SGLT-2 inhibitors reduced admission for heart to −0.54); low certainty) and GLP-1 receptor agonists
failure (odds ratio 0.70 (95% confidence interval 0.63 (−0.89% (−0.95 to −0.82); low certainty) might lower
to 0.77); 2, 9, 23, 29, and 58 fewer per 1000 in five glycated haemoglobin A1c levels more than placebo
years for very low, low, moderate, high, and very high (low certainty). GLP-1 receptor agonists reduced
risk patients respectively, moderate to high certainty; glycated haemoglobin A1c levels to a greater extent
appendix 8). GLP-1 receptor agonists had little or no than SGLT-2 inhibitors (mean difference −0.28% (95%
effect on hospitalisation for heart failure (0.94 (0.85 to CI −0.37 to −0.19); high certainty).
1.03), moderate to high certainty; appendix 8). SGLT-
2 inhibitors reduced hospitalisation for heart failure Other outcomes
compared with GLP-1 receptor agonists (0.74 (0.65 GLP-1 receptor agonists reduced serious
to 0.85); 1, 7, 18, 24, and 48 fewer per 1000 in five hyperglycaemia with moderate to high certainty (15
years for very low, low, moderate, high, and very high fewer events in 1000 patients treated with GLP-1
risk patients respectively, moderate to high certainty; receptor agonists over five years). Results showed the
table 3). effects of treatment on amputation or neuropathic pain
were very uncertain (appendix 8).
Blindness and eye disease requiring intervention
Seven trials including 68 221 patients reported Harms
blindness. The results of the network meta-analysis No difference in the odds of serious hypoglycaemia
were uninformative as blindness was rarely reported was found in a comparison of SGLT-2 inhibitor
and the confidence intervals for estimated treatment or GLP-1 receptor agonist treatment with placebo
effects were very wide (appendix 7). The outcome of or with each other in high or moderate certainty
eye disease requiring intervention was not specifically evidence (appendix 8). SGLT-2 inhibitor treatment
reported in any trial. and GLP-1 receptor agonist treatment probably did
not increase diabetic ketoacidosis compared with
Health related quality of life placebo or each other (moderate certainty). SGLT-2
Twenty five trials including 11 912 patients reported inhibitors increased genital infection compared with
health related quality of life during a median follow- placebo and GLP-1 receptor agonists (high certainty).
up of six months. The reported instruments included The effects of treatments on Fournier gangrene were
the Diabetes Treatment Satisfaction Questionnaire, the uncertain. GLP-1 receptor agonists might incur severe
Short Form 12 and 36, the EuroQoL five dimensions gastrointestinal events (low certainty). An increase

RESEARCH
in pancreatic cancer or pancreatitis is probably not endpoints, a finding not scientifically intuitive.
likely with SGLT-2 inhibitor or GLP-1 receptor agonist Trials generally did not include patients with lowest
treatment compared with placebo (low to moderate cardiovascular risk. Accordingly, the certainty of
certainty), and there might be no difference between evidence was graded down for the lowest risk patients
SGLT-2 inhibitor and GLP-1 receptor agonist treatment owing to indirectness. The panel agreed that a non-
for pancreatic cancer (low certainty). SGLT-2 inhibitor contextualised approach was appropriate to assess
treatment might incur a lower risk of pancreatitis than imprecision in estimates of effects to inform judgments
GLP-1 receptor agonist treatment (low certainty). about the certainty of evidence in the network meta-
No evidence was found that treatment effects differed analysis. We therefore did not adjudicate whether the
according to trial level characteristics, including trial range of treatment effects for each outcome was trivial,
duration, BMI, cardiovascular risk, and presence of small, moderate, or large; assess critical outcomes
chronic kidney disease (appendix 9). simultaneously; or make inferences about their
value relative to each other.31 The related BMJ Rapid
Discussion Recommendation will adjudicate the importance of
Principal findings the absolute effects of treatment and rate certainty
This network meta-analysis provides high certainty accordingly. This approach might lower the adjudicated
evidence that SGLT-2 inhibitors and GLP-1 receptor evidence certainty for several outcomes that inform
agonists, when added to other diabetes treatment, the BMJ Rapid Recommendations. Whether SGLT-2
reduce mortality, non-fatal myocardial infarction, inhibitors and GLP-1 receptor agonists given together
kidney failure, and serious hyperglycaemia. In the provide additional benefits compared with each
absence of head-to-head trials, we also found high treatment alone is not answered by this review. Notably,
certainty evidence for notable differences between the effects of treatment on glycated haemoglobin
SGLT-2 inhibitors and GLP-1 receptor agonists; SGLT- A1c and body weight were provided in trials with an
2 inhibitors reduced admission to hospital for heart average follow up of six months. Possibly, the relative
failure more than GLP-1 receptor agonists, whereas effects of the SGLT-2 inhibitor and GLP-1 receptor
GLP-1 receptor agonists reduced non-fatal stroke more agonist treatment on these outcomes might have been
than SGLT-2 inhibitors (which appeared to have no different in shorter duration trials than in the longer
effect). Differential effects of SGLT-2 inhibitors and therapeutic evaluations of cardiovascular outcomes.
GLP-1 receptor agonists on other patient important
outcomes included lower body weight and increased Comparisons with other studies
quality of life, albeit with small absolute differences. This systematic review included substantially more
Harms also differed. SGLT-2 inhibitors caused trials—including large and recently published trials—
genital infection, and GLP-1 receptor agonists might and patients than previously reported reviews and is
increase severe gastrointestinal events. Importantly, based on priority setting and received oversight by a
the absolute benefits for cardiovascular and renal panel with a range of clinical and lived experiences.
outcomes varied substantially for patients, depending SGLT-2 inhibitors and GLP-1 receptor agonists had
on their absolute cardiovascular risk (very low similar treatment effects on mortality as observed by
to very high; https://magicevidence.org/match- a previous network meta-analysis published in 2016,
it/200820dist/#!/) Taken together, the results have although precision in the treatment estimates has
clear implications for clinical practice, as reflected in increased substantially.11 The results provide evidence
the BMJ Rapid Recommendations directly informed by to support guideline recommendations made in 2019
our systematic review. that patients at highest risk of cardiovascular disease
and kidney disease are likely to have important benefits
Strength and limitations of this study on risks of cardiovascular events and heart failure with
The strengths of the review include a sensitive SGLT-2 inhibitors.2 9 This network meta-analysis aims
search to identify eligible trials, and independent to provide a broader representation of both relative
study identification, selection, data extraction, and absolute estimates of a wide range of important
and adjudication of risk of bias by two reviewers. clinical outcomes.
The review used the GRADE approach to report the The analysis is consistent with large scale
certainty of available evidence and reported estimated observational analyses of SGLT-2 inhibitor treatment
absolute risks for all outcomes, including patients with in a range of countries and settings, which suggest
varying risks of cardiovascular and kidney disease. beneficial effects on a range of clinical outcomes,
Limitations include the heterogeneity in clinical including reduced mortality and heart failure in people
settings of the included trials, although the consistency with diabetes at low cardiovascular risk and those
of results across studies diminishes this concern. Some with atherosclerotic cardiovascular disease.118 119 The
outcomes resulted in imprecise estimates of effects and effects of treatment in the available randomised trials in
low certainty evidence. our analyses are smaller than seen with observational
data, as the trial data enable analyses that reduce
Clinical uncertainties the risks of confounding by treatment indication;
SGLT-2 inhibitors failed to reduce non-fatal stroke in in other words, in real world datasets, people who
the same way that they reduced other cardiovascular receive drug treatment might be healthier or have

RESEARCH
other characteristics that lead to better outcomes than receptor agonists. The findings of this research paper
people who are not prescribed treatment. The present raise questions about how they should be used in
review can discern important differences between the clinical practice. In the linked article, readers will
effect of SGLT-2 inhibitors and GLP-1 receptor agonists find recommendations on SGLT-2 inhibitors and GLP-
on mortality and heart failure, not identified in large 1 receptor agonists based on data from this paper. To
scale non-randomised studies.120 do this a guideline panel has considered how direct
this evidence is and using GRADE methodology has
Policy implications integrated this with, for example, the values and
This systematic review of 764 clinical trials provides preferences of patients and resource implications. To
detailed information for decision makers about the read more about the guidelines please see the guideline
benefits and harms of SGLT-2 inhibitors and GLP-1 article in this package.
receptor agonists on important outcomes in adults
AUTHOR AFFILIATIONS
with type 2 diabetes. It includes trials with publicly 1
Department of Medicine, University of Otago, Christchurch, New
available information up to August 2020. A core Zealand
finding suggests that there is high certainty that 2
School of Public Health and Preventive Medicine, Monash
SGLT-2 inhibitors reduce all cause and cardiovascular University, Melbourne, VIC, Australia
3
mortality, non-fatal myocardial infarction, kidney Department of Internal Medicine, Division of Nephrology and
Hypertension, University of Kansas, Kansas City, KS, USA
failure, and admission to hospital for heart failure. 4
Department of Health Research Methods, Evidence and Impact,
GLP-1 receptor agonist treatment reduces all cause McMaster University, Hamilton, ON, Canada
and cardiovascular mortality, non-fatal myocardial 5
Institute of Health and Society, University of Oslo, Oslo, Norway
infarction, non-fatal stroke, and kidney failure. SGLT-2 6
Department of Endocrinology and Metabolism, West China
inhibitor treatment reduces admission to hospital for Hospital, Sichuan University, Chengdu, China
7
heart failure to a greater extent than GLP-1 receptor Division of Population Health and Genomics, Ninewells Hospital,
University of Dundee, Dundee, UK
agonist treatment, whereas GLP-1 receptor agonist 8
Centre for Gerontology and Geriatrics, West China Hospital,
treatment reduces non-fatal stroke more than SGLT-2
Sichuan University, Chengdu, China
inhibitors. The effect of SGLT-2 inhibitors and GLP- 9
Sydney School of Public Health, University of Sydney, Sydney, NSW,
1 receptor agonists on many other important patient Australia
outcomes is uncertain. Glycated haemoglobin A1c 10
Department of Emergency and Organ Transplantation, University
was not chosen as an important outcome in the review of Bari, Piazza Giulio CESARE, 70124 Bari, Italy
11
by the oversight panel, which included patients and Centre for Outcomes Research and Clinical Epidemiology
(CORESEARCH), Pescara, Italy
clinicians, but it was reported as it is considered by 12
Department of Nephrology, Division of Medicine, University of
some decision makers to be a measure of glycaemic Queensland at Princess Alexandra Hospital, Woolloongabba, QLD,
control and indicative of a more immediate measure of Australia
diabetes control. 13
Cumming School of Medicine, University of Calgary, Calgary, AB,
Canada
14
Conclusions George Institute for Global Health, Sydney, NSW, Australia
15
Trials of SGLT-2 inhibitors in people with existing Division of Nephrology, Hôpital Maisonneuve-Rosemont,
Montréal, QC, Canada
cardiovascular disease and chronic kidney disease with 16
Mater Private Clinic, Brisbane, QLD, Australia
or without diabetes are emerging or ongoing,7 6 9 121 17
Department of Nephrology, Faculty of Medicine and Dentistry,
indicating that SGLT-2 treatment decreases heart University of Alberta, Edmonton, AB, Canada
failure in the absence of type 2 diabetes. The findings of 18
Endocrine and Diabetes Unit, Queen Elizabeth Hospital,
the Dapagliflozin and Prevention of Adverse outcomes Woodville, SA, Australia
19
in Chronic Kidney Disease (DAPA-CKD) randomised Endocrine and Metabolic Unit, Royal Adelaide Hospital, Adelaide,
SA, Australia
controlled trial are awaited to identify whether 20
Garvan Institute of Medical Research, Sydney, NSW, Australia
dapagliflozin decreases kidney and cardiovascular 21
Department of Medicine, University of Melbourne, Melbourne, VIC,
events in people regardless of the presence of Australia
diabetes.121 22
Department of General Practice and Primary Health Care,
In conclusion, this network meta-analysis provides University of Melbourne, Melbourne, VIC, Australia
evidence of the absolute benefits and harms of SGLT- 23
Plataforma INVEST Medicina UANL-KER Unit Mayo Clinic (KER Unit
2 inhibitors and GLP-1 receptor agonists. The balance Mexico), Universidad Autonoma de Nuevo Leon, Monterrey, Mexico
24
of benefits and harms depends on individual risk Evidence-based Medicine Research Centre, Jiangxi University of
Traditional Chinese Medicine, Nanchang, China
profiles of patients, and thus the results support 25
Chinese Evidence-based Medicine Centre, Cochrane China Centre
a risk stratified approach for provision of SGLT-2 26
West China School of Medicine, Sichuan University, Chengdu,
inhibitors and GLP-1 receptor agonists to patients China
with type 2 diabetes. Accordingly, the associated BMJ 27
Faculty of Medicine and Health, Charles Perkins Centre, University
Rapid Recommendations will provide risk stratified of Sydney, Sydney, NSW, Australia
recommendations and highlight the need for shared 28
Department of Medicine, McMaster University, Hamilton, ON,
decision making to allow patients and clinicians to Canada
29
make well informed decisions together. Population Health Research Institute, Hamilton Health Sciences,
McMaster University, Hamilton, ON, Canada
These data provide direct evidence for the absolute
We thank members of the Rapid Recommendations panel for critical
benefits and harms of SGLT-2 inhibitors and GLP-1 feedback on the review question and outcome selection, GRADE

RESEARCH
judgments, and manuscript. We thank Ruth Mitchell for developing 4 Marso SP, Daniels GH, Brown-Frandsen K, et al, LEADER Steering
the search strategy. Committee, LEADER Trial Investigators. Liraglutide and cardiovascular
outcomes in type 2 diabetes. N Engl J Med 2016;375:311-22.
Contributors: SCP, AN, and GFMS conceived the study. SCP and GFMS
doi:10.1056/NEJMoa1603827
designed the search strategy. SCP performed the literature search. 5 Wiviott SD, Raz I, Bonaca MP, et al, DECLARE–TIMI 58 Investigators.
SCP, BT, MR, PN, VS, DWJ, MCR, SVB, YC, ACNF, MB, LIF, AL, NA, YL, Dapagliflozin and cardiovascular outcomes in type 2 diabetes.
ST, TM, LG, NK, RDB, RM, AKRC, HW, XYC, XZ, JL, AFR, ADGC, YW, LL, N Engl J Med 2019;380:347-57. doi:10.1056/NEJMoa1812389
SS, RCS, and FDG screened studies for eligibility; SCP, MR, and PN 6 Hernandez AF, Green JB, Janmohamed S, et al, Harmony Outcomes
assessed the risk of bias; and SCP, BT, MR, PN, VS, DWJ, SVB, YC, ACNF, committees and investigators. Albiglutide and cardiovascular outcomes
MB, LIF, AL, NA, YL, ST, TM, LG, NK, RB, RM, AKRC, HW, XC, XZ, JL, AFR, in patients with type 2 diabetes and cardiovascular disease (Harmony
ADGC, YW, LL, SS, RCS, and FDG performed data extraction. SCP, BT, Outcomes): a double-blind, randomised placebo-controlled trial.
RAM, POV, SL, QH, AN, DT, DWJ, MT, MW, GG, and GFMS interpreted Lancet 2018;392:1519-29. doi:10.1016/S0140-6736(18)32261-X
the data analysis; SCP, MR, PN, and GFMS assessed the certainty of 7 McGuire DK, Cannon CP, Pratley R, et al. Design and baseline
the evidence; SCP wrote the first draft of the manuscript; and all other characteristics of the eValuation of ERTugliflozin effIcacy and
authors revised the manuscript. SCP and GFMS are guarantors. The Safety CardioVascular outcomes trial (VERTIS-CV). Diabetologia
corresponding author attests that all listed authors meet authorship 2018;61(Supplement 1):S305-6.
criteria and that no others meeting the criteria have been omitted. 8 Perkovic V, Jardine MJ, Neal B, et al, CREDENCE Trial Investigators.
Canagliflozin and renal outcomes in type 2 diabetes and
Funding: This systematic review did not receive any funding. nephropathy. N Engl J Med 2019;380:2295-306. doi:10.1056/
Competing interests: All authors have completed the ICMJE uniform NEJMoa1811744
disclosure form at www.icmje.org/coi_disclosure.pdf and declare: SCP, 9 Cosentino F, Grant PJ, Aboyans V, et al, ESC Scientific Document
BT, RAM, POV, SL, QH, DT, MR, PN, VS, YC, ACNF, MB, LIF, AL, NA, YL, ST, Group. 2019 ESC Guidelines on diabetes, pre-diabetes, and
TM, NK, RDB, RM, AKRC, HW, XC, XZ, JL, AFR, ADGC, YW, LL, SS, RCS, cardiovascular diseases developed in collaboration with the EASD.
FDG, RRG, MW, GG, GFMS: no support from any organisation for the Eur Heart J 2020;41:255-323. doi:10.1093/eurheartj/ehz486
submitted work; no financial relationships with any organisations that 10 National Institute for Health and Care Excellence. Type 2 diabetes
might have an interest in the submitted work in the previous three in adults: management (NICE Guideline No. 28). Retrieved
years. SL was supported by grants from the National Natural Science from https://www.nice.org.uk/guidance/ng28/chapter/1-
Recommendations#drug-treatment-2. 2019.
Foundation of China (grant number 21534008), Sichuan Science and
11 Palmer SC, Mavridis D, Nicolucci A, et al. Comparison of clinical
Technology Programme (grant number 2019YFH0150), and 1.3.5
outcomes and adverse events associated with glucose lowering
Project for Disciplines of Excellence, West China Hospital, Sichuan drugs in patients with type 2 diabetes: a meta-analysis.
University (grant numbers ZYGD18022 and 2020HXF011). None of JAMA 2016;316:313-24. doi:10.1001/jama.2016.9400
these grants contribute to this work. AN reports grants and personal 12 Zheng SL, Roddick AJ, Aghar-Jaffar R, et al. Association between
fees from Novo Nordisk, grants from Sanofi, grants and personal use of sodium-glucose cotransporter 2 inhibitors, glucagon-like
fees from Astra Zeneca, grants from Pikdare, grants from AlfaSigma, peptide 1 agonists, and dipeptidyl peptidase 4 inhibitors with
outside the submitted work. DWJ reports grants and personal fees all-cause mortality in patients with type 2 diabetes: a systematic
from Baxter Healthcare, grants and personal fees from Fresenius review and meta-analysis. JAMA 2018;319:1580-91. doi:10.1001/
Medical Care, other from Amgen, personal fees from Astra Zeneca, jama.2018.3024
personal fees from AWAK, grants from National Health and Medical 13 Neuen BL, Young T, Heerspink HJL, et al. SGLT2 inhibitors for the
Research Council of Australia, personal fees from Ono, outside the prevention of kidney failure in patients with type 2 diabetes: a
submitted work. MT reports a grant to the institution by Daichi Sankyo systematic review and meta-analysis. Lancet Diabetes Endocrinol
in lieu of a personal honorarium. MCR reports grants from Sanofi, 2019;7:845-54. doi:10.1016/S2213-8587(19)30256-6
grants from NovoNordisk, grants from AstraZeneca, grants from 14 Hussein H, Zaccardi F, Khunti K, Seidu S, Davies MJ, Gray LJ.
Pikdare, during the conduct of the study. SVB reports advisory board Cardiovascular efficacy and safety of sodium-glucose co-
membership to Bayer Australia and AstraZeneca, speaking honoraria transporter-2 inhibitors and glucagon-like peptide-1 receptor
from Bayer Australia and Pfizer Australia; non-financial research agonists: a systematic review and network meta-analysis. Diabet
Med 2019;36:444-52. doi:10.1111/dme.13898
support from Bayer AG. LG reports personal fees from Boehringer-
15 Arnott C, Li Q, Kang A, et al. Sodium-glucose cotransporter 2
Engelheim, personal fees from Lilly, personal fees from Astra Zeneca,
inhibition for the prevention of cardiovascular events in patients with
non-financial support from Novo Nordisk, outside the submitted work. type 2 diabetes mellitus: a systematic review and meta-analysis. J Am
Ethical approval: Not required. Heart Assoc 2020;9:e014908.
Data sharing: No additional data available. 16 Rosenstock J, Perkovic V, Johansen OE, et al, CARMELINA
Investigators. Effect of linagliptin vs placebo on major cardiovascular
The manuscript’s guarantors (SCP and GFMS) affirm that the events in adults with type 2 diabetes and high cardiovascular
manuscript is an honest, accurate and transparent account of the and renal risk: the CARMELINA randomized clinical trial.
study being reported; that no important aspects of the study have JAMA 2019;321:69-79. doi:10.1001/jama.2018.18269
been omitted; and that any discrepancies from the study as planned 17 Siemieniuk RA, Agoritsas T, Macdonald H, Guyatt GH, Brandt
and registered have been explained. L, Vandvik PO. Introduction to BMJ Rapid Recommendations.
Dissemination to participants and related patient and public BMJ 2016;354:i5191. doi:10.1136/bmj.i5191
18 Higgins JP, Altman DG, Gøtzsche PC, et al, Cochrane Bias Methods
communities: The paper informs an upcoming BMJ Rapid
Group, Cochrane Statistical Methods Group. The Cochrane
Recommendation (https://www.bmj.com/rapid-recommendations)
Collaboration’s tool for assessing risk of bias in randomised trials.
on the use of SGLT-2 inhibitors and GLP-1 receptor agonists, also BMJ 2011;343:d5928. doi:10.1136/bmj.d5928
available in MAGICapp (https://www.magicapp.org/) for organisations 19 Liu Y, Wang W, Zhang AB, Bai X, Zhang S. Epley and Semont
to reuse or adapt for their own materials and purposes. maneuvers for posterior canal benign paroxysmal positional
Provenance and peer review: Not commissioned; externally peer vertigo: A network meta-analysis. Laryngoscope 2016;126:951-5.
reviewed. doi:10.1002/lary.25688
20 Veroniki AA, Vasiliadis HS, Higgins JP, Salanti G. Evaluation
This is an Open Access article distributed in accordance with the
of inconsistency in networks of interventions. Int J
Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, Epidemiol 2013;42:332-45. doi:10.1093/ije/dys222
which permits others to distribute, remix, adapt, build upon this work 21 Dias S, Welton NJ, Caldwell DM, Ades AE. Checking consistency in
non-commercially, and license their derivative works on different mixed treatment comparison meta-analysis. Stat Med 2010;29:932-
terms, provided the original work is properly cited and the use is non- 44. doi:10.1002/sim.3767
commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ 22 Furukawa TA, Barbui C, Cipriani A, Brambilla P, Watanabe N.
Imputing missing standard deviations in meta-analyses can provide
1 World Health Organization. Global Health Observatory (GHO) data. accurate results. J Clin Epidemiol 2006;59:7-10. doi:10.1016/j.
https://www.who.int/data/gho/data/themes/topics/causes-of-death. jclinepi.2005.06.006
World Health Organization, Geneva, Switzerland. 23 Hirji I, Andersson SW, Guo Z, Hammar N, Gomez-Caminero A.
2 American Diabetes Association. 9. Pharmacologic approaches to Incidence of genital infection among patients with type 2 diabetes in
glycemic treatment: standards of medical care in diabetes-2019. the UK General Practice Research Database. J Diabetes Complications
Diabetes Care 2019;42(Suppl 1):S90-102. doi:10.2337/dc19-S009 2012;26:501-5. doi:10.1016/j.jdiacomp.2012.06.012
3 Zinman B, Wanner C, Lachin JM, et al, EMPA-REG OUTCOME 24 Dieleman JP, Kerklaan J, Huygen FJ, Bouma PA, Sturkenboom MC.
Investigators. Empagliflozin, cardiovascular outcomes, and mortality Incidence rates and treatment of neuropathic pain conditions in
in type 2 diabetes. N Engl J Med 2015;373:2117-28. doi:10.1056/ the general population. Pain 2008;137:681-8. doi:10.1016/j.
NEJMoa1504720 pain.2008.03.002

RESEARCH
25 Takeuchi M, Kawamura T, Sato I, Kawakami K. Population-based 45 Hirano M, Nakamura T, Obata JE, et al. Early improvement in carotid
incidence of diabetic ketoacidosis in type 2 diabetes: medical claims plaque echogenicity by acarbose in patients with acute coronary
data analysis in Japan. Pharmacoepidemiol Drug Saf 2018;27:123- syndromes. Circ J 2012;76:1452-60. doi:10.1253/circj.CJ-11-1524
6. doi:10.1002/pds.4271 46 Hong SJ, Choi SC, Cho JY, et al. Pioglitazone increases circulating
26 Hippisley-Cox J, Coupland C. Development and validation of risk microRNA-24 with decrease in coronary neointimal hyperplasia in
prediction equations to estimate future risk of blindness and type 2 diabetic patients- optical coherence tomography analysis. Circ
lower limb amputation in patients with diabetes: cohort study. J 2015;79:880-8. doi:10.1253/circj.CJ-14-0964
BMJ 2015;351:h5441. doi:10.1136/bmj.h5441 47 Kaku K, Daida H, Kashiwagi A, et al. Long-term effects of pioglitazone
27 Basu S, Sussman JB, Berkowitz SA, Hayward RA, Yudkin JS. in Japanese patients with type 2 diabetes without a recent history
Development and validation of Risk Equations for Complications Of of macrovascular morbidity. Curr Med Res Opin 2009;25:2925-32.
type 2 Diabetes (RECODe) using individual participant data from doi:10.1185/03007990903328124
randomised trials. Lancet Diabetes Endocrinol 2017;5:788-98. 48 Laberge A, Brassard P, Arsenault B, et al. Positive effect of the PPAR-
doi:10.1016/S2213-8587(17)30221-8 gamma agonist rosiglitazone on hemodynamic response to exercise
28 Elley CR, Robinson T, Moyes SA, et al. Derivation and validation in type 2 diabetic men after coronary artery bypass graft surgery: a
of a renal risk score for people with type 2 diabetes. Diabetes 1-yr randomized study. Can J Cardiol 2016;32(Supplement 1):S223-
Care 2013;36:3113-20. doi:10.2337/dc13-0190 4. doi:10.1016/j.cjca.2016.07.356
29 Chaimani A, Higgins JP, Mavridis D, Spyridonos P, Salanti G. Graphical 49 Lee HW, Lee HC, Kim BW, et al. Effects of low dose pioglitazone
tools for network meta-analysis in STATA. PLoS One 2013;8:e76654. on restenosis and coronary atherosclerosis in diabetic patients
doi:10.1371/journal.pone.0076654 undergoing drug eluting stent implantation. Yonsei Med
30 Brignardello-Petersen R, Bonner A, Alexander PE, et al, GRADE J 2013;54:1313-20. doi:10.3349/ymj.2013.54.6.1313
Working Group. Advances in the GRADE approach to rate the 50 Moriwaki K, Takeuchi T, Fujimoto N, et al. Effect of sitagliptin on
certainty in estimates from a network meta-analysis. J Clin coronary flow reserve assessed by magnetic resonance imaging
Epidemiol 2018;93:36-44. doi:10.1016/j.jclinepi.2017.10.005 in type 2 diabetic patients with coronary artery disease. Circ
31 Hultcrantz M, Rind D, Akl EA, et al. The GRADE Working Group clarifies J 2018;82:2119-27. doi:10.1253/circj.CJ-18-0083
the construct of certainty of evidence. J Clin Epidemiol 2017;87:4-13. 51 Nishio K, Sakurai M, Kusuyama T, et al. A randomized comparison of
doi:10.1016/j.jclinepi.2017.05.006 pioglitazone to inhibit restenosis after coronary stenting in patients
32 Chaimai A, Caldwell DM, Li T, Higgins J, Salanti G. Chapter 11: with type 2 diabetes. Diabetes Care 2006;29:101-6. doi:10.2337/
Undertaking network meta-analyses. In: Higgins J, Thomas J, diacare.29.01.06.dc05-1170
Chandler J, et al, eds. Cochrane Handbook for Systematic Reviews 52 Ogasawara D, Shite J, Shinke T, et al. Pioglitazone reduces the
of Interventions version 61. 2020. www.training.cochrane.org/ necrotic-core component in coronary plaque in association with
handbook2020. enhanced plasma adiponectin in patients with type 2 diabetes
33 Gerstein HC, Ratner RE, Cannon CP, et al, APPROACH Study Group. mellitus. Circ J 2009;73:343-51. doi:10.1253/circj.CJ-08-0699
Effect of rosiglitazone on progression of coronary atherosclerosis in 53 Osman A, Otero J, Brizolara A, et al. Effect of rosiglitazone on
patients with type 2 diabetes mellitus and coronary artery disease: restenosis after coronary stenting in patients with type 2 diabetes.
the assessment on the prevention of progression by rosiglitazone Am Heart J 2004;147:e23. doi:10.1016/j.ahj.2003.12.006
on atherosclerosis in diabetes patients with cardiovascular 54 Nissen SE, Nicholls SJ, Wolski K, et al, PERISCOPE Investigators.
history trial. Circulation 2010;121:1176-87. doi:10.1161/ Comparison of pioglitazone vs glimepiride on progression of
CIRCULATIONAHA.109.881003 coronary atherosclerosis in patients with type 2 diabetes: the
34 Hedblad B, Zambanini A, Nilsson P, Janzon L, Berglund G. PERISCOPE randomized controlled trial. JAMA 2008;299:1561-73.
Rosiglitazone and carotid IMT progression rate in a mixed cohort of doi:10.1001/jama.299.13.1561
patients with type 2 diabetes and the insulin resistance syndrome: 55 Phrommintikul A, Wongcharoen W, Kumfu S, et al. Effects of
main results from the Rosiglitazone Atherosclerosis Study. J Intern dapagliflozin vs vildagliptin on cardiometabolic parameters in
Med 2007;261:293-305. doi:10.1111/j.1365-2796.2007.01767.x diabetic patients with coronary artery disease: a randomised study.
35 Cefalu WT, Leiter LA, de Bruin TW, Gause-Nilsson I, Sugg J, Parikh SJ. Br J Clin Pharmacol 2019;85:1337-47. doi:10.1111/bcp.13903
Dapagliflozin’s effects on glycemia and cardiovascular risk factors 56 Takagi T, Okura H, Kobayashi Y, et al, POPPS Investigators. A
in high-risk patients with type 2 diabetes: a 24-week, multicenter, prospective, multicenter, randomized trial to assess efficacy
randomized, double-blind, placebo-controlled study with a 28-week of pioglitazone on in-stent neointimal suppression in type 2
extension. Diabetes Care 2015;38:1218-27. doi:10.2337/dc14-0315 diabetes: POPPS (Prevention of In-Stent Neointimal Proliferation
36 Choi D, Kim SK, Choi SH, et al. Preventative effects of rosiglitazone by Pioglitazone Study). JACC Cardiovasc Interv 2009;2:524-31.
on restenosis after coronary stent implantation in patients with doi:10.1016/j.jcin.2009.04.007
type 2 diabetes. Diabetes Care 2004;27:2654-60. doi:10.2337/ 57 Tanaka A, Komukai S, Shibata Y, et al, Pioglitazone Reduce
diacare.27.11.2654 Inflammation and Restenosis with and without Drug Eluting Stent
37 Pfeffer MA, Claggett B, Diaz R, et al, ELIXA Investigators. Lixisenatide (PRIDE) Study Investigators. Effect of pioglitazone on cardiometabolic
in patients with type 2 diabetes and acute coronary syndrome. N profiles and safety in patients with type 2 diabetes undergoing
Engl J Med 2015;373:2247-57. doi:10.1056/NEJMoa1509225 percutaneous coronary artery intervention: a prospective,
38 Tanaka A, Shimabukuro M, Machii N, et al, EMBLEM Investigators. multicenter, randomized trial. Heart Vessels 2018;33:965-77.
Effect of empagliflozin on endothelial function in patients with type doi:10.1007/s00380-018-1143-3
2 diabetes and cardiovascular disease: results from the multicenter, 58 Dormandy JA, Charbonnel B, Eckland DJ, et al, PROactive
randomized, placebo- controlled, double-blind EMBLEM trial. Investigators. Secondary prevention of macrovascular events in
Diabetes Care 2019;42:e159-61. doi:10.2337/dc19-1177 patients with type 2 diabetes in the PROactive Study (PROspective
39 Verma S, Mazer CD, Yan AT, et al. Effect of empagliflozin on left pioglitAzone Clinical Trial In macroVascular Events): a randomised
ventricular mass in patients with type 2 diabetes and coronary controlled trial. Lancet 2005;366:1279-89. doi:10.1016/S0140-
artery disease: the EMPA-HEART CardioLink-6 randomized 6736(05)67528-9
clinical trial. Circulation 2019;140:1693-702. doi:10.1161/ 59 Hong J, Zhang Y, Lai S, et al, SPREAD-DIMCAD Investigators. Effects
CIRCULATIONAHA.119.042375 of metformin versus glipizide on cardiovascular outcomes in
40 Kuramitsu S, Miyauchi K, Yokoi H, et al. Effect of sitagliptin on plaque patients with type 2 diabetes and coronary artery disease. Diabetes
changes in coronary artery following acute coronary syndrome in Care 2013;36:1304-11. doi:10.2337/dc12-0719
diabetic patients: the ESPECIAL-ACS study. J Cardiol 2017;69:369- 60 Suzuki K, Tanaka S, Aoki C, Kato K, Jojima T, Aso Y. Greater efficacy
76. doi:10.1016/j.jjcc.2016.08.011 and improved endothelial dysfunction in untreated type 2 diabetes
41 White WB, Cannon CP, Heller SR, et al, EXAMINE Investigators. Alogliptin with liraglutide versus sitagliptin. Dokkyo J Med Sci 2014;41:211-20.
after acute coronary syndrome in patients with type 2 diabetes. N Engl 61 Kato Y, Iwata A, Zhang B, et al. Effects of dipeptidyl peptidase-4
J Med 2013;369:1327-35. doi:10.1056/NEJMoa1305889 inhibitor sitagliptin on coronary atherosclerosis as assessed by
42 Finn AV, Oh JS, Hendricks M, et al. Predictive factors for in-stent intravascular ultrasound in type 2 diabetes mellitus with coronary
late loss and coronary lesion progression in patients with type 2 artery disease. IJC Metab Endocr 2017;16:1-9. doi:10.1016/j.
diabetes mellitus randomized to rosiglitazone or placebo. Am Heart ijcme.2017.06.005
J 2009;157:383.e1-8. doi:10.1016/j.ahj.2008.11.013 62 Varghese A, Yee MS, Chan CF, et al. Effect of rosiglitazone
43 Gantz I, Chen M, Suryawanshi S, et al. A randomized, placebo-controlled on progression of atherosclerosis: insights using 3D carotid
study of the cardiovascular safety of the once-weekly DPP-4 inhibitor cardiovascular magnetic resonance. J Cardiovasc Magn
omarigliptin in patients with type 2 diabetes mellitus. Cardiovasc Reson 2009;11:24. doi:10.1186/1532-429X-11-24
Diabetol 2017;16:112. doi:10.1186/s12933-017-0593-8 63 Wang G, Wei J, Guan Y, Jin N, Mao J, Wang X. Peroxisome proliferator-
44 Hirano M, Nakamura T, Kitta Y, et al. Rapid improvement of carotid activated receptor-gamma agonist rosiglitazone reduces clinical
plaque echogenicity within 1 month of pioglitazone treatment inflammatory responses in type 2 diabetes with coronary artery
in patients with acute coronary syndrome. Atherosclerosis disease after coronary angioplasty. Metabolism 2005;54:590-7.
2009;203:483-8. doi:10.1016/j.atherosclerosis.2008.07.023 doi:10.1016/j.metabol.2004.11.017

RESEARCH
64 Wang Q, Wang D, Cheng A, Sun FY, Li Z. Comparison between the J Hypertens 2006;24:2047-55. doi:10.1097/01.
effects of sitagliptin and liraglutide on blood glucose and cognitive hjh.0000244955.39491.88
function of patients with both type 2 diabetes mellitus and post-stroke 83 Chacra A, Gantz I, Mendizabal G, et al. A randomised, double-blind,
mild cognitive impairment. Int J Clin Exp Med 2020;13:1219-27. trial of the safety and efficacy of omarigliptin (a once-weekly DPP-4
65 Liu B, Wang J, Wang G. Beneficial effects of pioglitazone on retardation inhibitor) in subjects with type 2 diabetes and renal impairment. Int J
of persistent atrial fibrillation progression in diabetes mellitus Clin Pract 2017;71:e12955. doi:10.1111/ijcp.12955
patients. Int Heart J 2014;55:499-505. doi:10.1536/ihj.14-107 84 Pollock C, Stefánsson B, Reyner D, et al. Albuminuria-lowering effect
66 Oe H, Nakamura K, Kihara H, et al, FESC, for Effect of a DPP-4 of dapagliflozin alone and in combination with saxagliptin and effect
inhibitor on left ventricular diastolic dysfunction in patients with type of dapagliflozin and saxagliptin on glycaemic control in patients
2 diabetes and diabetic cardiomyopathy (3D) study investigators. with type 2 diabetes and chronic kidney disease (DELIGHT): a
Comparison of effects of sitagliptin and voglibose on left ventricular randomised, double-blind, placebo-controlled trial. Lancet Diabetes
diastolic dysfunction in patients with type 2 diabetes: results of the Endocrinol 2019;7:429-41. doi:10.1016/S2213-8587(19)30086-
3D trial. Cardiovasc Diabetol 2015;14:83. doi:10.1186/s12933- 5
015-0242-z 85 Fioretto P, Del Prato S, Buse JB, et al, DERIVE Study Investigators.
67 Arturi F, Succurro E, Miceli S, et al. Liraglutide improves cardiac Efficacy and safety of dapagliflozin in patients with type 2 diabetes
function in patients with type 2 diabetes and chronic heart failure. and moderate renal impairment (chronic kidney disease stage
Endocrine 2017;57:464-73. doi:10.1007/s12020-016-1166-4 3A): the DERIVE Study. Diabetes Obes Metab 2018;20:2532-40.
68 Tanaka A, Hisauchi I, Taguchi I, et al, CANDLE Trial Investigators. doi:10.1111/dom.13413
Effects of canagliflozin in patients with type 2 diabetes and 86 Yale JF, Bakris G, Cariou B, et al, DIA3004 Study Group. Efficacy
chronic heart failure: a randomized trial (CANDLE). ESC Heart and safety of canagliflozin over 52 weeks in patients with type
Fail 2020;7:1585-94. doi:10.1002/ehf2.12707 2 diabetes mellitus and chronic kidney disease. Diabetes Obes
69 McMurray JJV, Solomon SD, Inzucchi SE, et al, DAPA-HF Trial Metab 2014;16:1016-27. doi:10.1111/dom.12348
Committees and Investigators. Dapagliflozin in patients with heart 87 Barnett AH, Mithal A, Manassie J, et al, EMPA-REG RENAL trial
failure and reduced ejection fraction. N Engl J Med 2019;381:1995- investigators. Efficacy and safety of empagliflozin added to existing
2008. doi:10.1056/NEJMoa1911303 antidiabetes treatment in patients with type 2 diabetes and chronic
70 Dargie HJ, Hildebrandt PR, Riegger GA, et al. A randomized, kidney disease: a randomised, double-blind, placebo-controlled trial.
placebo-controlled trial assessing the effects of rosiglitazone on Lancet Diabetes Endocrinol 2014;2:369-84. doi:10.1016/S2213-
echocardiographic function and cardiac status in type 2 diabetic 8587(13)70208-0
patients with New York Heart Association functional class I or II 88 Hiramatsu T, Asano Y, Mabuchi M, Imai K, Iguchi D, Furuta S.
heart failure. J Am Coll Cardiol 2007;49:1696-704. doi:10.1016/j. Liraglutide relieves cardiac dilated function than DPP-4 inhibitors. Eur
jacc.2006.10.077 J Clin Invest 2018;48:e13007. doi:10.1111/eci.13007
71 Sharma A, Ambrosy AP, DeVore AD, et al. Liraglutide and weight loss 89 Hu ZY, Zhao JW. Effect of rosiglitazone on serum CRP and plasma
among patients with advanced heart failure and a reduced ejection PAI-1 in patients with early type 2 diabetic nephropathy. Zhongyuan
fraction: insights from the FIGHT trial. ESC Heart Fail 2018;5:1035- Yikan 2007;34:27-8.
43. doi:10.1002/ehf2.12334 90 Ito M, Abe M, Okada K, et al. The dipeptidyl peptidase-4 (DPP-4)
72 Giles TD, Miller AB, Elkayam U, Bhattacharya M, Perez A. Pioglitazone inhibitor vildagliptin improves glycemic control in type 2 diabetic
and heart failure: results from a controlled study in patients patients undergoing hemodialysis. Endocr J 2011;58:979-87.
with type 2 diabetes mellitus and systolic dysfunction. J Card doi:10.1507/endocrj.EJ11-0025
Fail 2008;14:445-52. doi:10.1016/j.cardfail.2008.02.007 91 Jerums G, Murray RM, Seeman E, et al. Lack of effect of gliclazide on
73 Singh JSS, Mordi IR, Vickneson K, et al. Dapagliflozin versus placebo early diabetic nephropathy and retinopathy: a two-year controlled
on left ventricular remodeling in patients with diabetes and study. Diabetes Res Clin Pract 1987;3:71-80. doi:10.1016/S0168-
heart failure: the REFORM Trial. Diabetes Care 2020;43:1356-9. 8227(87)80010-4
doi:10.2337/dc19-2187 92 Kohan DE, Fioretto P, Tang W, List JF. Long-term study of patients
74 McMurray JJV, Ponikowski P, Bolli GB, et al, VIVIDD Trial Committees with type 2 diabetes and moderate renal impairment shows that
and Investigators. Effects of vildagliptin on ventricular function in dapagliflozin reduces weight and blood pressure but does not
patients with type 2 diabetes mellitus and heart failure: a randomized improve glycemic control. Kidney Int 2014;85:962-71. doi:10.1038/
placebo-controlled trial. JACC Heart Fail 2018;6:8-17. doi:10.1016/j. ki.2013.356
jchf.2017.08.004 93 Kashiwagi A, Takahashi H, Ishikawa H, et al. A randomized, double-
75 Abe M, Okada K, Maruyama T, Maruyama N, Soma M, Matsumoto blind, placebo-controlled study on long-term efficacy and safety of
K. Clinical effectiveness and safety evaluation of long-term ipragliflozin treatment in patients with type 2 diabetes mellitus and
pioglitazone treatment for erythropoietin responsiveness and insulin renal impairment: results of the long-term ASP1941 safety evaluation
resistance in type 2 diabetic patients on hemodialysis. Expert Opin in patients with type 2 diabetes with renal impairment (LANTERN)
Pharmacother 2010;11:1611-20. doi:10.1517/14656566.2010.4 study. Diabetes Obes Metab 2015;17:152-60. doi:10.1111/
95119 dom.12403
76 Abe M, Higuchi T, Moriuchi M, et al. Efficacy and safety of saxagliptin, 94 Davies MJ, Bain SC, Atkin SL, et al. Efficacy and safety of liraglutide
a dipeptidyl peptidase-4 inhibitor, in hemodialysis patients with versus placebo as add-on to glucose-lowering therapy in patients
diabetic nephropathy: a randomized open-label prospective with type 2 diabetes and moderate renal impairment (LIRA-RENAL): a
trial. Diabetes Res Clin Pract 2016;116:244-52. doi:10.1016/j. randomized clinical trial. Diabetes Care 2016;39:222-30.
diabres.2016.04.034 95 Lukashevich V, Schweizer A, Shao Q, Groop PH, Kothny W. Safety
77 Allegretti AS, Zhang W, Zhou W, et al. Safety and effectiveness of and efficacy of vildagliptin versus placebo in patients with type 2
bexagliflozin in patients with type 2 diabetes mellitus and stage diabetes and moderate or severe renal impairment: a prospective
3a/3b CKD. Am J Kidney Dis 2019;74:328-37. doi:10.1053/j. 24-week randomized placebo-controlled trial. Diabetes Obes
ajkd.2019.03.417 Metab 2011;13:947-54. doi:10.1111/j.1463-1326.2011.01467.x
78 Morikawa A, Ishizeki K, Iwashima Y, et al. Pioglitazone reduces urinary 96 Groop PH, Cooper ME, Perkovic V, et al. Linagliptin and its effects on
albumin excretion in renin-angiotensin system inhibitor-treated type hyperglycaemia and albuminuria in patients with type 2 diabetes and
2 diabetic patients with hypertension and microalbuminuria: the renal dysfunction: the randomized MARLINA-T2D trial. Diabetes Obes
APRIME study. Clin Exp Nephrol 2011;15:848-53. doi:10.1007/ Metab 2017;19:1610-9. doi:10.1111/dom.13041
s10157-011-0512-3 97 Maruyama T, Takashima H, Tei R, Furukawa T, Maruyama
79 Arjona Ferreira JC, Corry D, Mogensen CE, et al. Efficacy and safety N, Abe M. MON-300 Efficacy and safety of canagliflozin, a
of sitagliptin in patients with type 2 diabetes and ESRD receiving sodium glucose cotransporter 2 (SGLT2) inhibitor, in diabetic
dialysis: a 54-week randomized trial. Am J Kidney Dis 2013;61:579- kidney disease: a randomized open-label prospective trial.
87. doi:10.1053/j.ajkd.2012.11.043 Kidney Int Rep 2019;4(Supplement):S422-3. doi:10.1016/j.
80 Arjona Ferreira JC, Marre M, Barzilai N, et al. Efficacy and safety ekir.2019.05.1109
of sitagliptin versus glipizide in patients with type 2 diabetes 98 McGill JB, Sloan L, Newman J, et al. Long-term efficacy and safety
and moderate-to-severe chronic renal insufficiency. Diabetes of linagliptin in patients with type 2 diabetes and severe renal
Care 2013;36:1067-73. doi:10.2337/dc12-1365 impairment: a 1-year, randomized, double-blind, placebo-controlled
81 Tuttle KR, Lakshmanan MC, Rayner B, et al. Dulaglutide versus insulin study. Diabetes Care 2013;36:237-44. doi:10.2337/dc12-0706
glargine in patients with type 2 diabetes and moderate-to-severe 99 Milovanova L, Taranova M, Milovanova S, et al. DPP-4 inhibitors
chronic kidney disease (AWARD-7): a multicentre, open-label, (linagliptin) impact on klotho serum level in patients with type 2
randomised trial. Lancet Diabetes Endocrinol 2018;6:605-17. diabetes. Diabetes Technol Ther 2019;21(Supplement 1):A147.
doi:10.1016/S2213-8587(18)30104-9 100 Nakamura T, Ushiyama C, Osada S, Hara M, Shimada N, Koide H.
82 Bakris GL, Ruilope LM, McMorn SO, et al. Rosiglitazone reduces Pioglitazone reduces urinary podocyte excretion in type 2 diabetes
microalbuminuria and blood pressure independently of patients with microalbuminuria. Metabolism 2001;50:1193-6.
glycemia in type 2 diabetes patients with microalbuminuria. doi:10.1053/meta.2001.26703

RESEARCH
101 Nakamura T, Matsuda T, Kawagoe Y, et al. Effect of pioglitazone outcomes in patients with type 2 diabetes: the CAROLINA
on carotid intima-media thickness and arterial stiffness in type randomized clinical trial. JAMA 2019;322:1155-66. doi:10.1001/
2 diabetic nephropathy patients. Metabolism 2004;53:1382-6. jama.2019.13772
doi:10.1016/j.metabol.2004.05.013 113 Ikonomidis I, Pavlidis G, Thymis J, et al. Effects of glucagon-like
102 Nakamura T, Sugaya T, Kawagoe Y, Ueda Y, Koide H. Effect of peptide-1 receptor agonists, sodium-glucose cotransporter-2
pioglitazone on urinary liver-type fatty acid-binding protein inhibitors, and their combination on endothelial glycocalyx,
concentrations in diabetes patients with microalbuminuria. Diabetes arterial function, and myocardial work index in patients with
Metab Res Rev 2006;22:385-9. doi:10.1002/dmrr.633 type 2 diabetes mellitus after 12-month treatment. J Am Heart
103 Neff KJ, Tobin LM, Hogan AE, Docherty NG, Le Roux CW, O’Shea D. Assoc 2020;9:e015716. doi:10.1161/JAHA.119.015716
The effect of low dose liraglutide on renal inflammation in type 2 114 Li J, Feng Z, Li Q, He Y, Zhao C, He J. Insulin glargine effectively
diabetic kidney disease: a randomised controlled study. Diabet achieves glycemic control and improves insulin resistance in
Med 2016;33(Suppl 1):64. patients with early type 2 diabetes that exhibit a high risk for
104 Nowicki M, Rychlik I, Haller H, et al. Long-term treatment with the cardiovascular disease. Exp Ther Med 2014;8:147-52. doi:10.3892/
dipeptidyl peptidase-4 inhibitor saxagliptin in patients with type 2 etm.2014.1688
diabetes mellitus and renal impairment: a randomised controlled 115 McGuire DK, Abdullah SM, See R, et al. Randomized comparison
52-week efficacy and safety study. Int J Clin Pract 2011;65:1230-9. of the effects of rosiglitazone vs. placebo on peak integrated
doi:10.1111/j.1742-1241.2011.02812.x cardiovascular performance, cardiac structure, and function. Eur
105 Pistrosch F, Passauer J, Herbrig K, Schwanebeck U, Gross P, Bornstein Heart J 2010;31:2262-70. doi:10.1093/eurheartj/ehq228
SR. Effect of thiazolidinedione treatment on proteinuria and renal 116 Scirica BM, Bhatt DL, Braunwald E, et al, SAVOR-TIMI 53 Steering
hemodynamic in type 2 diabetic patients with overt nephropathy. Committee and Investigators. Saxagliptin and cardiovascular
Horm Metab Res 2012;44:914-8. doi:10.1055/s-0032-1314836 outcomes in patients with type 2 diabetes mellitus. N Engl J
106 Raji A, Scott R, Morgan J, et al. Safety and efficacy of sitagliptin Med 2013;369:1317-26. doi:10.1056/NEJMoa1307684
compared with dapagliflozin in patients with type 2 diabetes, mild 117 Yee MS, Pavitt DV, Dhanjil S, Godsland IF, Richmond W, Johnston
renal impairment, and inadequate glycaemic control on metformin DG. The effects of rosiglitazone on atherosclerotic progression in
+/− a sulfonylurea. Diabetologia 2018;61(Supplement 1):S379-80. patients with type 2 diabetes at high cardiovascular risk. Diabet
107 Halden TAS, Kvitne KE, Midtvedt K, et al. Efficacy and safety of Med 2010;27:1392-400. doi:10.1111/j.1464-5491.2010.03089.x
empagliflozin in renal transplant recipients with posttransplant 118 Birkeland KI, Jørgensen ME, Carstensen B, et al. Cardiovascular
diabetes mellitus. Diabetes Care 2019;42:1067-74. doi:10.2337/ mortality and morbidity in patients with type 2 diabetes following
dc19-0093 initiation of sodium-glucose co-transporter-2 inhibitors versus
108 Takashima H, Yoshida Y, Nagura C, et al. Renoprotective effects of other glucose-lowering drugs (CVD-REAL Nordic): a multinational
canagliflozin, a sodium glucose cotransporter 2 inhibitor, in type observational analysis. Lancet Diabetes Endocrinol 2017;5:709-17.
2 diabetes patients with chronic kidney disease: a randomized doi:10.1016/S2213-8587(17)30258-9
open-label prospective trial. Diab Vasc Dis Res 2018;15:469-72. 119 Kosiborod M, Cavender MA, Fu AZ, et al, CVD-REAL Investigators
doi:10.1177/1479164118782872 and Study Group*. Lower risk of heart failure and death in patients
109 Grunberger G, Camp S, Johnson J, et al. Ertugliflozin in patients with initiated on sodium-glucose cotransporter-2 inhibitors versus
stage 3 chronic kidney disease and type 2 diabetes mellitus: the other glucose-lowering drugs. Circulation 2017;136:249-59.
VERTIS RENAL randomized study. Diabetes Ther 2018;9:49-66. doi:10.1161/CIRCULATIONAHA.117.029190
doi:10.1007/s13300-017-0337-5 120 Poonawalla I, Bowe A, Tindal M, Meah Y, Schwab P. 36-OR:
110 Yale JF, Bakris G, Cariou B, et al. Efficacy and safety of canagliflozin in Comparative effectiveness of SGLT2 inhibitors vs. GLP-1 agonists.
subjects with type 2 diabetes and chronic kidney disease. Diabetes Diabetes 2020;69(Supplement 1) doi:10.2337/db20-36-OR
Obes Metab 2013;15:463-73. doi:10.1111/dom.12090 121 Heerspink HJL, Stefansson BV, Chertow GM, et al, DAPA-CKD
111 Neal B, Perkovic V, Mahaffey KW, et al, CANVAS Program Collaborative Investigators. Rationale and protocol of the Dapagliflozin
Group. Canagliflozin and cardiovascular and renal events in And Prevention of Adverse outcomes in Chronic Kidney
type 2 diabetes. N Engl J Med 2017;377:644-57. doi:10.1056/ Disease (DAPA-CKD) randomized controlled trial. Nephrol Dial
NEJMoa1611925 Transplant 2020;35:274-82. doi:10.1093/ndt/gfz290
112 Rosenstock J, Kahn SE, Johansen OE, et al, CAROLINA Investigators.
Effect of linagliptin vs glimepiride on major adverse cardiovascular Web appendix: Supplementary material

SGLT 2 Inhibitor Dan GLP 1 RA

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SGLT 2 Inhibitor Dan GLP 1 RA

Uploaded by

Copyright:

Available Formats

RESEARCH

Sodium-glucose cotransporter protein-2 (SGLT-2) inhibitors and

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 1

2 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 3

4 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

Basal bolus insulin

Alpha glucosidase inhibitor Placebo

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 5

6 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 7

8 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 9

10 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 11

12 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 13

14 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

You might also like

SGLT 2 Inhibitor Dan GLP 1 RA

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SGLT 2 Inhibitor Dan GLP 1 RA

Uploaded by

Copyright:

Available Formats

RESEARCH

Sodium-glucose cotransporter protein-2 (SGLT-2) inhibitors and

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 1

2 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 3

4 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

Basal bolus insulin

Alpha glucosidase inhibitor Placebo

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 5

6 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 7

8 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 9

10 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 11

12 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 13

14 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

You might also like

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 1

2 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 3

4 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 5

6 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 7

8 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 9

10 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 11

12 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj

the bmj | BMJ 2021;372:m4573 | doi: 10.1136/bmj.m4573 13

14 doi: 10.1136/bmj.m4573 | BMJ 2021;372:m4573 | the bmj